I like this piece a lot, but its origin is as much a mystery to me as to you. The pile part looks Jaff Kurdish as hell, an impression reinforced by the offset knotting that Patrick Weiler pointed out. The "birds on a pole" are so ubiquitous in western and central Asia that their presence doesn't help much. The colors certainly look NW Persian or Caucasian (I mention both in order to not upset those who see NW Iran and Azerbaijan as really different places in the 19th century). One feature that catches my attention is the cotton warps. Very odd, indeed.

And, as you point out, not only is its origin mysterious, it doesn't match any of the familiar formats in terms of what it was used for.

Steve Price

Steve

I took a quick look (well, a couple of hours, really) at all of the Hali magazines from issue 50 on. Also the majority of the ORR magazines. When I have seen bag faces similar to the one Jerry has presented, they have been designated as Kurdish. My impressions were that they look a bit Bakhtiari, but probably only because of the mix of techniques. The one thought I had each time I saw one was that I had not seen one in print that I could recall. My cursory survey of Hali and ORR confirms that this type of weaving is either rare enough to not even have been published recently, or that it has been determined insignificant enough to be not worth printing. (Jerry, I am sure you will pick the former)
A review of Eiland (newly updated) shows a similar configuration, plate 76 - Kurdish bagface, with two panels of weaving, but both panels are pile; the top half of which is standard symmetrical knots but the bottom part is done with offset knotting.
The authors say that the older Kurdish pieces have wool foundations, but later pieces have more cotton. On the topic of offset knotting, they indicate that, although the Jaf Kurds are known to use offset knotting, it's presence in Anatolian Kurdish rugs does not determine a Jaf weaving.
One conclusion that is likely pertinent to this weaving is: "Clearly the designs are influenced by proximity to major weaving areas. Kurds living closest to Anatolia tend to produce rugs with a Turkish flavor, while those from the extreme northern Kurdish areas are more likely to absorb some Caucasian influence."
This statement would indicate a Kurdish weaver close to a Shahsavan area may have woven this curious piece. Probably a member of the Jafsavan tribe

Patrick Weiler

I do not understand the connection you are making between plate 76 in Eiland and Jerry's piece. Jerry's piece measures 44 inches horizontally and 30 inches vertically. Plate 76 measures 29 inches horizontally and 32 inches vertically (74 x 81 cm). I have had plate 76 in my hands. If you visit Mark Santos, you may have seen it as well. It is almost certainly a detached Khorjin face and is entirely done in knotted pile unlike Jerry's piece.

The only remarkable thing about plate 76 is that the curvilinear part of the face (which was probably folded under the end of the bag when complete) has no offset knotting while the design part with the hexagonal elements does have offset knotting. Neither of the designs is particularly unusual. The format with a kind of skirt (here rather large and with a Herati design) is also not that uncommon in these bags. What is unusual is this combination of designs and the large size of the skirt. Otherwise, the piece is fairly standard - and it has two ply warps to the best of my recollection.

Thanks, Michael

You are correct, there is very little similar about the two weavings other than they are both 2-section weavings with different designs in the top and bottom sections and that one of the sections of weave has offset knots.

And that is the only point I was making, that there just aren't many, if any, examples like Jerry's that I am aware of in the literature. The closest thing I could find in over 50 Hali magazines was a picture of a Sanjabi Kurd chuval face on page 102 of issue 103. It is even farther removed than the Eiland weaving from being similar to Jerry's example. It appears to be fully pile woven and the lower part (elem?) is smaller compared to the top than Jerry's. It does, however, have what I remember to be a more typical (typical of the few of these I have seen) top section layout of two panels of the same design but of different colors. The examples I have seen would have a top section of soumak weave with a design like that of the Hali example and a pile lower section. I also do not remember any closure tabs or loops on any of these weavings.

You mentioned Mark Santos. I saw the first example of Jerry's type of bagface at his gallery several years ago, with a top soumak section and bottom pile section. It was something I had not seen before and only seen a couple of times since. Could this mean it was made only by a small group of weavers, or for a function not common enough to warrant more widespread use? The Edsel comes to mind Is it possible that more of these existed but were cut in half when sold? That would make two pieces of decorated weavings, except the pile section on Jerry's is an odd size not found on the market. Steve mentions above that it does not fit the standard format of functional items more commonly available. Both sections are fully decorated, unlike most bags with the back rather plain.

I was expecting a plethora of postings with photographs of similar, if not identical, weavings from the extensive collections of Tukotek correspondents. I have heard about a forthcoming book on Kurdish weavings. We may have to wait a while to find out what this thing Jerry has really is.

Patrick Weiler

There is a large consensus to admit that this piece is NW Persian and I proposed a possible Lori/Varamin attribution where Lori tribes have migrated from SW Persia.

Let me try again.

Jerry’s storage sack face has three main characteristics which normally would help us to determine its origin.
1/ Very large size
2/ one soumak panel and one pile bottom
3/ large use of offset knotting

In NW Persia:
Which tribes have woven such large storage sacks: to my knowledge, only Jaf, Sanjabi and Lori-Bakhtiary tribes.
Which tribes have woven “large” storage sacks with a soumak panel and an pile bottom: To my knowledge only Lori-Bakhtiary.
Which tribes used extensively offset knotting: To my knowledge only Jaf Kurds. I have seen offset knotting in the borders of Lori weavings to add bulk in these area but not any rug with a large use of offset knotting.

So probabilities are 2/3 for a Jaf or a Lori origin. Can someone else do better?

Patrick,
I looked closely to plate 76 in Eiland’s book. The largest pile panel contains hooked polygons with a rendering which is normally associated with South Eastern Anatolian weavings. In Eastern Anatolia offset knotting is not rare and I pointed recently on Ebay an Eastern Anatolian rug with the Baklava design. All the design was woven with offset knots and Marla confirmed my observation.

Thanks,
Daniel

Looking at the book, Tribal Rugs, by Jim Opie, there is a well known Luri bag front,plate 4.30 with dimensions of 45" wide by 22" high, with a narrow pile band at the bottom. If this pile band were larger, the measurements would approximate the bag Jerry has shown. The Opie bag is first half nineteenth century. He also shows a Bakhtiyari bag front in figure 8.12 that consists of a sumak top and pile bottom. There is no dimensional or structural information, though, but it seems narrower and the top section is taller than Jerry's. (This was one of the shortcomings noted in the Hali review of the book, if I remember correctly)
Regarding Kurds, Opie notes: "Close similarities to old Luri and Bakhtiyari nomadic motifs is a noteworthy feature of some of these designs."
So, Daniel, the construction of this bag, with offset knots, argues more for a Kurdish manufacture with Luri influenced design, although the size and layout is more similar to a "standard" (for a type with very few members) Luri format. I have a couple of Jaf Kurd bags that are large, one 36x20 and the other 40x30, but both are pile only, with offset knots.

Unequivocally confused now,

Patrick Weiler

I just recall that we showed the photo of large Sanjabi Kurd torba during my Salon 42 which is illustrated in O'Bannon's book Tribal and Village Rugs from Arizona Collections - plate 45. Sizes of this one are VERY LARGE, 4Ǝ" x V3ƌ" and one of the two panel was woven in SOUMAK technique while the other one was knotted.
Please read back the comments of several participants in the "archieved Salon discussions ".

Sanjabi weavers may be also used ofsett knotting.

It appears more and more to me that this piece has been woven by Sanjabi or Jaf Kurds as we know for sure that they woven such large storage sacks, used ofsett knotting to shape the design of their rugs and as this example shows it have sometimes woven one panel in soumak technique.

I think we must not be far now, from the exact origin of this piece.

Best,
Daniel

Part of the problem in making an attribution here may be that Jerry's piece is sufficiently late enough to fall partly out of any tradition known to collectors. Whether it is or not, I think we are pushing too far for an answer.

To date, I haven't seen any attribution made in this Salon that fit very comfortably. Whether the warps are three ply or machine spun cotton, they are not what I would expect of a Kurdish weaving. I have not seen anyone link the selvedge to what we typically find on Kurdish weavings.

Regarding Sanjabi Kurd weavings, one of the points O'Bannon made in his tentative attempt to label this group is that their weavings have a slightly different color palette and do not use offset knots. So how exactly would this piece fit there? Offset knots seem to be the one fact that everyone is relying on to force a Kurdish attribution on this weaving. Maybe this is correct, maybe it is not. But I think we may be over emphasizing it.

I also do not accept the argument that what we have is a Kurdish weaving with Luri influence ala Opie's theory. In my mind Daniel's initial thought of Luri/Varamin area did not seem too forced. There does seem to be a SW Persian influence in the piece. More troubling to me is accepting Opie's thesis that Kurds just adapted Bahktiary and Luri tribal motifs - it only takes us to a chicken and the egg question and is too dependent on extrapolating from the existence of Luristan bronzes. What are those Luri and Bahktiary motifs and who created them first? This is a picture that is very unclear and extremely complicated. I think Marla would have a lot to say about the origin of these motifs that has nothing at all to do with tribe.

I do not think we are any closer to an attribution than we were. It may just be that this weaving does not fit into a category for any number of reasons. Perhaps what we have is a village product.

Regrets, Michael